Radiotelephone voice control device, in particular for use in a motor vehicle

Abstract

The apparatus comprises a data memory containing a series of correspondents' call numbers and, for each call number, at least one associated voice print; a sound transducer suitable for picking up the name of a desired corespondent as spoken by the user of the apparatus; voice recognition means suitable for analyzing the correspondent's name as picked up by the transducer and for transforming it into an associated voice print; selective memory addressing means including associative means suitable for finding a voice print in the memory corresponding to the print supplied by the voice recognition means, and in the event of a match, for addressing the corresponding memory position; and means co-operating with the associative means for applying the addressed call number to the radiotelephone circuits. The voice recognition means evaluate and store a current noise level as picked up by the transducer in the absence of a speech signal; when in the presence of a speech signal, they subtract the previously evaluated current noise level from the signal as picked up; and then they apply the resulting signal as obtained in this way to a DTW type voice recognition algorithm with pattern recognition by dynamic programming adapted to speech using dynamic parameter extraction functions, in particular a predictive dynamic algorithm with forward and/or backward and/or frequency masking.

What is claimed is:  
     
       1. Voice control apparatus for a radiotelephone, in particular for use in a motor vehicle, the apparatus comprising: 
       a data memory containing a series of correspondents' call numbers and, for each call number, at least one associated voice print;  
       a sound transducer suitable for picking up the name of a desired correspondent as uttered by the user of the apparatus;  
       voice recognition means suitable for analyzing the correspondent's name as picked up by the transducer and for transforming it into an associated voice print;  
       selective memory addressing means including associative means suitable for finding a voice print in the memory corresponding to the print supplied by the voice recognition means, and in the event of a match, for addressing the corresponding memory position; and  
       means cooperating with the associative means for applying the addressed call number to the radiotelephone circuits;  
       the voice recognition means comprising:  
       means capable of evaluating and storing a current noise level as picked up by the transducer in the absence of a speech signal, said noise being picked up continuously by the sound transducer and analyzed to give a weighted mean energy spectrum of the noise, which spectrum is continuously updated;  
       means capable, in the presence of a speech signal, of subtracting the previously evaluated noise power spectrum from the power spectrum of the signal as picked up; and  
       means capable of applying the resulting signal as obtained in this way to a DTW type voice recognition algorithm with pattern recognition by dynamic programming adapted to speech using dynamic parameter extraction functions, in particular a predictive dynamic algorithm with forward and/or backward and/or frequency masking, further comprising a key actuated by the user and whose actuation controls radiotelephone control functions including answering, hanging up, and swapping during call waiting.  
     
     
       2. Voice control apparatus for a radiotelephone, in particular for use in a motor vehicle, the apparatus comprising: 
       a data memory containing a series of correspondents' call numbers and, for each call number, at least one associated voice print;  
       a sound transducer suitable for picking up the name of a desired correspondent as uttered by the user of the apparatus;  
       voice recognition means suitable for analyzing the correspondent's name as picked up by the transducer and for transforming it into an associated voice print;  
       selective memory addressing means including associative means suitable for finding a voice print in the memory corresponding to the print supplied by the voice recognition means, and in the event of a match, for addressing the corresponding memory position; and  
       means cooperating with the associative means for applying the addressed call number to the radiotelephone circuits;  
       the voice recognition means comprising:  
       means capable of evaluating and storing a current noise level as picked up by the transducer in the absence of a speech signal, said noise being picked up continuously by the sound transducer and analyzed to give a weighted mean energy spectrum of the noise, which spectrum is continuously updated;  
       means capable, in the presence of a speech signal, of subtracting the previously evaluated noise power spectrum from the power spectrum of the signal as picked up; and  
       means capable of applying the resulting signal as obtained in this way to a DTW type voice recognition algorithm with pattern recognition by dynamic programming adapted to speech using dynamic parameter extraction functions, in particular a predictive dynamic algorithm with forward and/or backward and/or frequency masking, further comprising a key actuated by the user to mark a speech signal, and in which the signal picked up by the transducer is stored continuously over a given time interval (k 0 ) so as to perform voice recognition over a length of time ([T 0 −k 0 , T 1 +k 1 ]) including the period during which said key is actuated ([T 0 , T 1 ]) plus said time interval (k 0 ) immediately preceding said actuation.  
     
     
       3. Voice control apparatus for a radiotelephone, in particular for use in a motor vehicle, the apparatus comprising: 
       a data memory containing a series of correspondents' call numbers and, for each call number, at least one associated voice print;  
       a sound transducer suitable for picking up the name of a desired correspondent as uttered by the user of the apparatus;  
       voice recognition means suitable for analyzing the correspondent's name as picked up by the transducer and for transforming it into an associated voice print;  
       selective memory addressing means including associative means suitable for finding a voice print in the memory corresponding to the print supplied by the voice recognition means, and in the event of a match, for addressing the corresponding memory position; and  
       means cooperating with the associative means for applying the addressed call number to the radiotelephone circuits;  
       the voice recognition means comprising:  
       means capable of evaluating and storing a current noise level as picked up by the transducer in the absence of a speech signal, said noise being picked up continuously by the sound transducer and analyzed to give a weighted mean energy spectrum of the noise, which spectrum is continuously updated;  
       means capable, in the presence of a speech signal, of subtracting the previously evaluated noise power spectrum from the power spectrum of the signal as picked up; and  
       means capable of applying the resulting signal as obtained in this way to a DTW type voice recognition algorithm with pattern recognition by dynamic programming adapted to speech using dynamic parameter extraction functions, in particular a predictive dynamic algorithm with forward and/or backward and/or frequency masking, further comprising a key actuated by the user to mark a speech signal, and in which voice recognition is performed over a length of time ([T 0 −k 0 , T 1 +k 1 ]) terminating after the end (T 1 ) of actuation of the key.